Chien, Eric2020-09-082020-09-082020-06https://hdl.handle.net/11299/216089University of Minnesota M.S. thesis. June 2020. Major: Conservation Biology. Advisor: Susan Galatowitsch. 1 computer file (PDF); viii, 74 pages.Practitioners of ecological restoration are faced with making sense of a wide range of information and generating conclusions about recovery potential that can inform effort and planning. In this thesis I use resilience theory to provide a framework for generating insights about ecological dynamics of ecosystems that tie directly to fundamental restoration assessment challenges. The design and function of resilience assessments needs to be informed by theory-based concepts as well as the needs and practice constraints of practitioners. There are few examples of resilience-based site assessments for restoration despite the theoretical link between resilience and fundamental practical restoration questions. Mid-continental North American oak savanna is introduced as a model system in which to design, test, and validate a resilience assessment for restoration practice. The objective was to create a tool that could reliably assess current condition of prospective restorations and diagnose recovery potential and dynamics. Beginning with defining the range of possible intact and altered states for a given ecosystem, this resilience assessment uses practically measurable attributes to gauge the recovery potential of a site. Resilience attributes include those that confer resistance to likely disturbances and regeneration ability for both desired and altered state features. The tool was initially designed through a process of rapid prototyping using academic and management literature and broad scoping of oak savanna restoration practitioners and projects. The design process included parameterizing the tool using field data from 47 oak savanna restoration projects. A refined, fully operational tool was then field tested by new and experienced practitioners for its usability, consistency, and validity. Results from tool testing showed the tool effectively supported consistent conclusions about site condition and recovery. Relative to using unstructured narrative site assessment, tool-supported practitioners used an expanded quantity of ecological information in site assessment, and the tool facilitated transparent decision-making with collected information. The validation approach was hampered by a limited ability to compare conclusions between tool-supported practitioners and unstructured expert assessments, as well as a lack of available pre-restoration starting condition information to retroactively populate the entire assessment tool using restoration case histories. With modifications to the validation approach, the process documented for the design and testing of a resilience-based site assessment of oak savannas has potential to serve as an effective model for expanding the development of similar tools for the restoration of other terrestrial ecosystems.enoak savannaresiliencerestorationsite assessmentThesis or Dissertation